Dynamic creation of data specification-driven AI-based executable strategies for high availability of evaluation services

Abstract

A service provider may create data specification-driven AI-based executable strategies, which may be used to ensure high uptime and availability of decision services. A service provider, such as an electronic transaction processor for digital transactions, may utilize different decision services that implement rules and artificial intelligence models for decision-making of data including data in production computing environment. A decision service may normally be used for data processing and decision-making. However, at certain times, the decision service may fail or the services and/or a gateway for such services may be inaccessible. To provide higher availability and better SLA times, a client-side executable strategy for decision service execution may be determined using the pathways for strategy execution and available data from called resources. This strategy may be loaded in parallel to calling the decision service, and when failure occurs, may be used as a fallback to request processing.

What is claimed is: 
     
       1. A service provider system comprising:
 a non-transitory memory; and 
 one or more hardware processors coupled to the non-transitory memory and configured to read instructions from the non-transitory memory to cause the service provider system to perform operations comprising:
 parsing a strategy utilized for processing a data processing request from a computing device by a decision service of the service provider system, wherein the strategy comprises a plurality of pathways for an execution of the strategy, each of the plurality of pathways including one or more data processing nodes for required data by each of the plurality of pathways during the execution of the strategy; 
 determining, based on the parsing, the required data for each of the plurality of pathways; 
 determining available data to the decision service from one or more additional services that are callable using an application programming interface (API) specification associated with the decision service; 
 identifying a first pathway of the plurality of pathways available to be used for the strategy based on the available data and the required data for each of the plurality of pathways; 
 creating a new strategy executable client-side by the computing device during a failure associated with the decision service at the service provider system; 
 storing the new strategy in a data repository available to the computing device; 
 determining the failure of the decision service has occurred; 
 receiving, from the computing device, a result of processing the new strategy client-side by the computing device, wherein the result is determined using an artificial intelligence (AI) model for an application on the computing device; and 
 processing the data processing request by the decision service using at least the result. 
 
 
     
     
       2. The service provider system of  claim 1 , wherein the operations further comprise:
 determining a second pathway of the plurality of pathways available to be used for the strategy based on the available data and the required data, 
 wherein the new strategy is created to include the first pathway and the second pathway as alternate pathways for processing the new strategy by the computing device in place of the decision service processing the strategy. 
 
     
     
       3. The service provider system of  claim 1 , wherein the processing the data processing request by the decision service using at least the result comprises continuing with a pathway that executes the strategy after completing one or more nodes in the pathway using the result from the computing device. 
     
     
       4. The service provider system of  claim 1 , wherein the storing the new strategy comprises:
 packaging the new strategy in a data structure storable by the data repository, wherein the packaging includes stitching the first pathway with any second pathways available to be used for the strategy based on the available data and the required data. 
 
     
     
       5. The service provider system of  claim 1 , wherein the storing the new strategy is in response to one of an initiation of the decision service or a refresh of the decision service with an application on the computing device. 
     
     
       6. The service provider system of  claim 1 , wherein, prior to the determining the failure of the decision service has occurred, the operations further comprise:
 receiving the data processing request; 
 initiating a processing the data processing request using at least the strategy with the decision service prior to the failure of the decision service occurring, wherein the initiating the processing begins a first one of the plurality of pathways for processing the data processing request, and wherein at least one of the one or more data processing nodes in the first one of the plurality of pathways is completed by the result of processing the new strategy client-side after the failure has occurred. 
 
     
     
       7. The service provider system of  claim 1 , wherein the determining the available data comprises:
 executing one or more API calls to the one or more additional services using the API specification; 
 determining if any of the one or more API calls fails or is unresponsive; and 
 updating a list of callable services for the one or more additional services based on the determining if any of the one or more API calls fails or is unresponsive. 
 
     
     
       8. The service provider system of  claim 1 , wherein the determining the available data includes deriving at least a portion of the available data using the AI model of the service provider system from data available via calling the one or more additional services. 
     
     
       9. The service provider system of  claim 1 , wherein the decision service comprises a decision service or a microservice of a service-based application architecture for the service provider system. 
     
     
       10. A method comprising:
 determining, by parsing a first strategy executable by a decision service of a service provider, a plurality of pathways used when executing the first strategy by the decision service and required data for each of the plurality of pathways, wherein each of the plurality of pathways includes one or more data processing nodes of the required data; 
 determining available data from one or more computing resources that are callable using an application programming interface (API) specification associated with the decision service, wherein the available data is used by the one or more data processing nodes when executing the first strategy; 
 generating, based on the available data and the required data by each of the plurality of pathways, a second strategy from the first strategy, wherein the second strategy is executable by a plurality of computing devices client-side during a failure associated with the decision service, and wherein the client-side execution of the second strategy includes calling the second strategy in parallel to a call to the decision service of the service provider that executes the first strategy in response to a data processing request by one of the plurality of computing devices; and 
 storing the second strategy in a data repository accessible by a plurality of computing devices. 
 
     
     
       11. The method of  claim 10 , further comprising:
 creating, prior to the generating, the first strategy server-side by the service provider. 
 
     
     
       12. The method of  claim 10 , wherein the client-side execution of the second strategy includes determining the failure of the decision service of the service provider that executes the first strategy based on the call in parallel to the decision service. 
     
     
       13. The method of  claim 10 , wherein, prior to the generating the second strategy, the method further comprises:
 determining at least one available pathway of the plurality of pathways for executing the first strategy based on the available data and the required data by the one or more data processing nodes in each of the plurality of pathways, 
 wherein the second strategy is generated based further on the at least one available pathway. 
 
     
     
       14. The method of  claim 13 , wherein the at least one available pathway is determined based further on one or more corresponding data processing nodes using the available data, and wherein the available data is determined for a current time that the one or more computing resources are callable using the API specification. 
     
     
       15. The method of  claim 10 , wherein the one or more data processing nodes comprise components of the first strategy and that provide a processing result of the first strategy for requests by the plurality of computing devices. 
     
     
       16. A non-transitory machine-readable medium having stored thereon machine-readable instructions executable to cause a machine to perform operations comprising:
 parsing a strategy utilized for processing a data processing request from a computing device by a decision service of a service provider system, wherein the strategy comprises a plurality of pathways for an execution of the strategy, each of the plurality of pathways including one or more data processing nodes for required data by each of the plurality of pathways during the execution of the strategy; 
 determining, based on the parsing, the required data for each of the plurality of pathways; 
 determining available data to the decision service from one or more additional services that are callable using an application programming interface (API) specification associated with the decision service; 
 identifying a first pathway of the plurality of pathways available to be used for the strategy based on the available data and the required data for each of the plurality of pathways; 
 creating a new strategy executable client-side by the computing device during a failure associated with the decision service at the service provider system; 
 storing the new strategy in a data repository available to the computing device; 
 determining the failure of the decision service has occurred; 
 receiving, from the computing device, a result of processing the new strategy client-side by the computing device, wherein the result is determined based on an artificial intelligence (AI) model for an application on the computing device; and 
 processing the data processing request by the decision service using at least the result. 
 
     
     
       17. The non-transitory machine-readable medium of  claim 16 , wherein the operations further comprise:
 determining a second pathway of the plurality of pathways available to be used for the strategy based on the available data and the required data, 
 wherein the new strategy is created to include the first pathway and the second pathway as alternate pathways for processing the new strategy by the computing device in place of the decision service processing the strategy. 
 
     
     
       18. The non-transitory machine-readable medium of  claim 16 , wherein the processing the data processing request by the decision service using at least the result comprises continuing with a pathway that executes the strategy after completing one or more nodes in the pathway using the result from the computing device. 
     
     
       19. The non-transitory machine-readable medium of  claim 16 , wherein the storing the new strategy comprises:
 packaging the new strategy in a data structure storable by the data repository, wherein the packaging includes stitching the first pathway with any second pathways available to be used for the strategy based on the available data and the required data. 
 
     
     
       20. The non-transitory machine-readable medium of  claim 16 , wherein the storing the new strategy is in response to one of an initiation of the decision service or a refresh of the decision service with an application on the computing device.